Failure to Identify HIV-Infected Individuals in a Clinical Trial Using a Single HIV Rapid Test for Screening Estelle Piwowar-Manning,1 Jessica M. Fogel,1 Oliver Laeyendecker,2,3 Shauna Wolf,1 Vanessa Cummings,1 Mark A. Marzinke,1 William Clarke,1 Autumn Breaud,1 Sarah Wendel,3 Lei Wang,4 Priscilla Swanson,5 John Hackett, Jr,5 Sharon Mannheimer,6 Carlos del Rio,7 Irene Kuo,8 Nina T. Harawa,9 Beryl A. Koblin,10 Richard Moore,2 Joel N. Blankson,2 and Susan H. Eshleman1 1
Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland; 2Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland; 3Laboratory of Immunoregulation, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland; 4Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Research Center, Seattle, Washington; 5Infectious Disease Research, Abbott Diagnostics, Abbott Park, Illinois; 6Department of Medicine, Harlem Hospital, Columbia University, Mailman School of Public Health, New York, New York; 7Department of Global Health, Emory University Rollins School of Public Health, Atlanta, Georgia; 8Department of Epidemiology and Biostatistics, The George Washington University, Washington, DC; 9 Department of Research, Charles R. Drew University of Medicine and Science, Los Angeles, California; 10Laboratory of Infectious Disease Prevention, Lindsley F. Kimball Research Institute, New York Blood Center, New York, New York
Background: In the HIV Prevention Trials Network (HPTN) 061 study, 8 (2.3%) of 348 HIV-infected participants identified as HIV uninfected at study enrollment using a single HIV rapid test for screening were found to be HIV infected after additional testing. Objectives: To evaluate the performance of different HIV assays for detection of HIV infection in HPTN 061 participants with missed infection and individuals with viral suppression. Methods: Plasma samples from 8 HPTN 061 participants, 17 elite controllers, and 101 individuals on antiretroviral treatment (ART) were tested for HIV with 3 rapid tests, 2 laboratory-based immunoassays, and a Western blot assay. The HPTN 061 samples were also tested with 2 HIV RNA assays and an antiretroviral drug assay. Results: Of the 8 HPTN 061 participants with missed infection, 1 was an elite controller, 1 was taking ART, 2 were missed because of testing or clerical errors, 1 had recent HIV infection (identified using a multi-assay algorithm), and 3 had acute HIV infection. Two (1.7%) of 118 individuals with viral suppression (both taking ART) had at least 1 false-negative test. Conclusions: In clinical trials, HIV infections can be missed for a variety of reasons. Using more than one assay to screen for HIV infection may reduce the number of missed infections. Key words: antiretroviral therapy, elite controller, HIV, rapid test, viral suppression
IV rapid tests are commonly used to screen for HIV infection in clinical, community, and research settings. False-negative HIV rapid test results have been observed in some studies during the early stages of HIV infection.1,2 For example, one study evaluated the performance of 4 US Food and Drug Administration (FDA)–approved HIV rapid tests. In that study, most individuals who had detectable HIV RNA with a negative or indeterminate Western blot had nonreactive rapid test results.1 False-negative HIV rapid test results have also been observed in individuals with advanced disease3 and individuals receiving antiretroviral treatment (ART).4,5 In one study, 3 of 6 FDA-approved HIV rapid tests had at least one false-negative test result when samples from individuals on ART were analyzed.5 Failure
H
62
to identify HIV-infected individuals in clinical trials can confound study outcomes and can put those individuals at risk if the study includes an intervention, such as provision of antiretroviral drugs for pre-exposure prophylaxis (PrEP).6 Although HIV testing is not usually performed
Corresponding author: Susan Eshleman, MD, PhD, Department of Pathology, Johns Hopkins University School of Medicine, 646 Ross Building, 720 Rutland Avenue, Baltimore, MD 21205; phone: 410-614-4734; fax: 410-502-9244; e-mail: seshlem@ jhmi.edu HIV Clin Trials 2014;15(2):62–68 © 2014 Thomas Land Publishers, Inc. www.hivclinicaltrials.com doi: 10.1310/hct1502-62
FALSE-NEGATIVE HIV TESTS •
for individuals with a prior HIV diagnosis, it may occur in clinical settings or clinical trials to confirm self-reported HIV status or to determine HIV status in individuals who are aware of their HIV infection but choose not to disclose this information.7–9 The HIV Prevention Trials Network (HPTN) 061 study (NCT00951249) assessed the feasibility and acceptability of a multicomponent intervention for HIV prevention among Black men who have sex with men (MSM) in the United States.10,11 The study enrolled 1,553 men, including HIV-uninfected men, HIV-infected men who reported no prior HIV diagnosis, HIV-infected men who reported that they were HIV infected but not in care, and a limited number of HIV-infected men who were in care. Study participants were screened for HIV infection at enrollment at study sites using a single HIV rapid antibody test; testing was repeated 6 and 12 months after enrollment.10,11 Plasma samples were sent to a centralized laboratory for retrospective quality assurance testing. This retrospective HIV testing identified 8 HIV-infected men who had nonreactive HIV rapid tests among the 1,500 men who had HIV rapid testing performed at study enrollment. In this report, we analyzed samples from those 8 men to understand why their infections were missed. We also evaluated the impact of viral suppression on the performance of HIV screening assays by testing samples from a cohort of elite controllers and from HIV-infected adults on ART. METHODS Samples Used for Analysis Plasma samples were obtained from the 8 HPTN 061 participants described above, 17 elite controllers who were virally suppressed in the absence of ART (EC group),12,13 and 101 individuals who were virally suppressed from ART (ART group).14 HIV infection was diagnosed a median of 12 years prior to sample collection (interquartile range [IQR], 5–17 years) in the EC group and a median of 8 years prior to sample collection (IQR, 4–13 years) in the ART group. In the ART group, individuals were virally suppressed from ART for a median of 1.6 years prior to sample collection (IQR, 266 days to 6 years). Additional information for the EC and ART groups is provided in Table 1.
PIWOWAR-MANNING ET AL
63
Laboratory Testing Real-time HIV rapid testing was performed at HPTN 061 study sites with venous blood using the OraQuick Advance HIV-1/2 Antibody Test (OraSure Technologies, Bethlehem, PA). Retrospective testing of HPTN 061 samples and testing of samples from the EC and ART groups was performed at a centralized laboratory (HPTN Laboratory Center, Johns Hopkins University, Baltimore, MD) using 3 HIV rapid tests (OraQuick test; Uni-Gold Recombigen HIV Test, Trinity Biotech, Wicklow, Ireland; and INSTI Rapid HIV Test, BioLytical Laboratories, Richmond, BC, Canada). The samples were also analyzed using a thirdgeneration enzyme immunoassay (EIA, VITROS Anti-HIV 1+2 Test, Ortho Clinical Diagnostics, Rochester, NY), a fourth-generation chemiluminescent microparticle immunoassay (CMIA) (ARCHITECT HIV Ag/Ab Combo Assay [List 2P36], Abbott Laboratories, Wiesbaden, Germany), and a Western blot assay (Genetics System HIV-1 Western Blot, Bio-Rad Laboratories, Redmond, WA). Two samples from the ART group did not have sufficient volume to perform all assays. In addition, samples from HPTN 061 participants were tested with 2 HIV RNA assays: (1) either the AMPLICOR HIV-1 Ultrasensitive MONITOR Test, version 1.5 (14 samples) or the COBAS AmpliPrep/COBAS TaqMan HIV-1 Test, v2.0 (8 samples; Roche Molecular Diagnostics, Indianapolis, IN), and (2) the APTIMA HIV-1 RNA Qualitative Assay (Gen-Probe Inc., San Diego, CA). These assays have different window periods for HIV detection.15 Samples from HPTN 061 participants were also screened for the presence of antiretroviral (ARV) drugs, including 9 protease inhibitors (PIs), 4 nucleoside/nucleotide reverse transcriptase inhibitors, and 2 non-nucleoside reverse transcriptase inhibitors (NNRTIs).16 Detection of an NNRTI or PI was considered to be indicative of ART. The individuals who performed testing at the centralized laboratory were blinded to the HIV status of the HPTN 061 participants. Ethical Approval Written informed consent was obtained for participation in the HPTN 061 study. The study was approved by the institutional review boards at the participating institutions. Individuals in the
64
HIV CLINICAL TRIALS •
15/2
•
MAR-APR 2014
Table 1. Characteristics of virally suppressed study subjects Gender (% male)
Race (% African American)
Mean CD4 cell count (IQR), cells/mm3
Mean HIV viral load, copies/mL
Group
n
Median age at testing (IQR), years
EC
17
56 (53–58)
52.9
100.0
No
817 (732–1,141)
750,000 11,762 75,478 3,031f 53,148 34
Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland; 2Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland; 3Laboratory of Immunoregulation, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland; 4Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Research Center, Seattle, Washington; 5Infectious Disease Research, Abbott Diagnostics, Abbott Park, Illinois; 6Department of Medicine, Harlem Hospital, Columbia University, Mailman School of Public Health, New York, New York; 7Department of Global Health, Emory University Rollins School of Public Health, Atlanta, Georgia; 8Department of Epidemiology and Biostatistics, The George Washington University, Washington, DC; 9 Department of Research, Charles R. Drew University of Medicine and Science, Los Angeles, California; 10Laboratory of Infectious Disease Prevention, Lindsley F. Kimball Research Institute, New York Blood Center, New York, New York
Background: In the HIV Prevention Trials Network (HPTN) 061 study, 8 (2.3%) of 348 HIV-infected participants identified as HIV uninfected at study enrollment using a single HIV rapid test for screening were found to be HIV infected after additional testing. Objectives: To evaluate the performance of different HIV assays for detection of HIV infection in HPTN 061 participants with missed infection and individuals with viral suppression. Methods: Plasma samples from 8 HPTN 061 participants, 17 elite controllers, and 101 individuals on antiretroviral treatment (ART) were tested for HIV with 3 rapid tests, 2 laboratory-based immunoassays, and a Western blot assay. The HPTN 061 samples were also tested with 2 HIV RNA assays and an antiretroviral drug assay. Results: Of the 8 HPTN 061 participants with missed infection, 1 was an elite controller, 1 was taking ART, 2 were missed because of testing or clerical errors, 1 had recent HIV infection (identified using a multi-assay algorithm), and 3 had acute HIV infection. Two (1.7%) of 118 individuals with viral suppression (both taking ART) had at least 1 false-negative test. Conclusions: In clinical trials, HIV infections can be missed for a variety of reasons. Using more than one assay to screen for HIV infection may reduce the number of missed infections. Key words: antiretroviral therapy, elite controller, HIV, rapid test, viral suppression
IV rapid tests are commonly used to screen for HIV infection in clinical, community, and research settings. False-negative HIV rapid test results have been observed in some studies during the early stages of HIV infection.1,2 For example, one study evaluated the performance of 4 US Food and Drug Administration (FDA)–approved HIV rapid tests. In that study, most individuals who had detectable HIV RNA with a negative or indeterminate Western blot had nonreactive rapid test results.1 False-negative HIV rapid test results have also been observed in individuals with advanced disease3 and individuals receiving antiretroviral treatment (ART).4,5 In one study, 3 of 6 FDA-approved HIV rapid tests had at least one false-negative test result when samples from individuals on ART were analyzed.5 Failure
H
62
to identify HIV-infected individuals in clinical trials can confound study outcomes and can put those individuals at risk if the study includes an intervention, such as provision of antiretroviral drugs for pre-exposure prophylaxis (PrEP).6 Although HIV testing is not usually performed
Corresponding author: Susan Eshleman, MD, PhD, Department of Pathology, Johns Hopkins University School of Medicine, 646 Ross Building, 720 Rutland Avenue, Baltimore, MD 21205; phone: 410-614-4734; fax: 410-502-9244; e-mail: seshlem@ jhmi.edu HIV Clin Trials 2014;15(2):62–68 © 2014 Thomas Land Publishers, Inc. www.hivclinicaltrials.com doi: 10.1310/hct1502-62
FALSE-NEGATIVE HIV TESTS •
for individuals with a prior HIV diagnosis, it may occur in clinical settings or clinical trials to confirm self-reported HIV status or to determine HIV status in individuals who are aware of their HIV infection but choose not to disclose this information.7–9 The HIV Prevention Trials Network (HPTN) 061 study (NCT00951249) assessed the feasibility and acceptability of a multicomponent intervention for HIV prevention among Black men who have sex with men (MSM) in the United States.10,11 The study enrolled 1,553 men, including HIV-uninfected men, HIV-infected men who reported no prior HIV diagnosis, HIV-infected men who reported that they were HIV infected but not in care, and a limited number of HIV-infected men who were in care. Study participants were screened for HIV infection at enrollment at study sites using a single HIV rapid antibody test; testing was repeated 6 and 12 months after enrollment.10,11 Plasma samples were sent to a centralized laboratory for retrospective quality assurance testing. This retrospective HIV testing identified 8 HIV-infected men who had nonreactive HIV rapid tests among the 1,500 men who had HIV rapid testing performed at study enrollment. In this report, we analyzed samples from those 8 men to understand why their infections were missed. We also evaluated the impact of viral suppression on the performance of HIV screening assays by testing samples from a cohort of elite controllers and from HIV-infected adults on ART. METHODS Samples Used for Analysis Plasma samples were obtained from the 8 HPTN 061 participants described above, 17 elite controllers who were virally suppressed in the absence of ART (EC group),12,13 and 101 individuals who were virally suppressed from ART (ART group).14 HIV infection was diagnosed a median of 12 years prior to sample collection (interquartile range [IQR], 5–17 years) in the EC group and a median of 8 years prior to sample collection (IQR, 4–13 years) in the ART group. In the ART group, individuals were virally suppressed from ART for a median of 1.6 years prior to sample collection (IQR, 266 days to 6 years). Additional information for the EC and ART groups is provided in Table 1.
PIWOWAR-MANNING ET AL
63
Laboratory Testing Real-time HIV rapid testing was performed at HPTN 061 study sites with venous blood using the OraQuick Advance HIV-1/2 Antibody Test (OraSure Technologies, Bethlehem, PA). Retrospective testing of HPTN 061 samples and testing of samples from the EC and ART groups was performed at a centralized laboratory (HPTN Laboratory Center, Johns Hopkins University, Baltimore, MD) using 3 HIV rapid tests (OraQuick test; Uni-Gold Recombigen HIV Test, Trinity Biotech, Wicklow, Ireland; and INSTI Rapid HIV Test, BioLytical Laboratories, Richmond, BC, Canada). The samples were also analyzed using a thirdgeneration enzyme immunoassay (EIA, VITROS Anti-HIV 1+2 Test, Ortho Clinical Diagnostics, Rochester, NY), a fourth-generation chemiluminescent microparticle immunoassay (CMIA) (ARCHITECT HIV Ag/Ab Combo Assay [List 2P36], Abbott Laboratories, Wiesbaden, Germany), and a Western blot assay (Genetics System HIV-1 Western Blot, Bio-Rad Laboratories, Redmond, WA). Two samples from the ART group did not have sufficient volume to perform all assays. In addition, samples from HPTN 061 participants were tested with 2 HIV RNA assays: (1) either the AMPLICOR HIV-1 Ultrasensitive MONITOR Test, version 1.5 (14 samples) or the COBAS AmpliPrep/COBAS TaqMan HIV-1 Test, v2.0 (8 samples; Roche Molecular Diagnostics, Indianapolis, IN), and (2) the APTIMA HIV-1 RNA Qualitative Assay (Gen-Probe Inc., San Diego, CA). These assays have different window periods for HIV detection.15 Samples from HPTN 061 participants were also screened for the presence of antiretroviral (ARV) drugs, including 9 protease inhibitors (PIs), 4 nucleoside/nucleotide reverse transcriptase inhibitors, and 2 non-nucleoside reverse transcriptase inhibitors (NNRTIs).16 Detection of an NNRTI or PI was considered to be indicative of ART. The individuals who performed testing at the centralized laboratory were blinded to the HIV status of the HPTN 061 participants. Ethical Approval Written informed consent was obtained for participation in the HPTN 061 study. The study was approved by the institutional review boards at the participating institutions. Individuals in the
64
HIV CLINICAL TRIALS •
15/2
•
MAR-APR 2014
Table 1. Characteristics of virally suppressed study subjects Gender (% male)
Race (% African American)
Mean CD4 cell count (IQR), cells/mm3
Mean HIV viral load, copies/mL
Group
n
Median age at testing (IQR), years
EC
17
56 (53–58)
52.9
100.0
No
817 (732–1,141)
750,000 11,762 75,478 3,031f 53,148 34
